Hepatitis C virus (HCV) is a highly variable RNA virus that has infected 185 million worldwide and is associated with severe liver diseases and cancer. While a successful vaccine may require robust T and B cell responses, the focus of this application is on a B cell-based vaccine. The vaccine must overcome the high diversity of HCV isolates and its potential for viral escape, and the vaccine must induce sterilizing immunity despite the low immunogenicity of epitopes that mediate broad virus neutralization (Vn). We have developed a unique database on the E2 glycoprotein, the natural target of the majority of broadly neutralizing (Vn) antibodies. The database outlines the immunogenic regions on E2 that segregate into five clusters of overlapping epitopes, designated antigenic domains A-E, and hypervariable region 1 (HVR1). Domain A elicits non-virus-neutralizing (non-Vn) antibodies and HVR1 elicits isolate-specific antibodies from which HCV can easily escape. By contrast, epitopes within domains B, D and E elicit broad Vn antibodies. We will exploit this database, as well as evidence that epitope conformational stability is a key determinant of immunogenicity and that Vn epitopes can be grafted onto heterologous protein scaffolds, to address the following questions: Does increasing the conformational stability of Vn B, D and E epitopes, or grafting them onto heterologous scaffolds, increase their antigenicity? Can E2 be glycoengineered to silence domain A decoy epitopes and up-modulate Vn epitopes? How can this information be employed to design an effective preventive vaccine? Our Aims are: 1. Structure- guided computational redesign of the E2 glycoprotein. We will use X-ray crystallographic information on human monoclonal antibodies (HMAbs) bound to B, D and E epitopes to computationally redesign E2 to stabilize these broadly Vn epitopes and thereby increase their immunogenicity. Conversely, we will silence domain A epitopes by strategic introduction of N-glycans. We will evaluate our most promising combinations of designed E2 mutants in the context of the E1E2 heterodimer. 2. Design of scaffolded E2 immunogens to optimize presentation of Vn epitopes. As a complementary approach to E2 redesign, we will engineer small scaffolded immunogens to elicit Vn antibodies to B, D and E epitopes. 3. Self-assembling polyphosphazene nanoparticles for multimeric E2 and E1E2 presentation. To achieve more effective presentation of designed immunogens from Aims 1 and 2, we will develop biodegradable synthetic nanoparticles to display them in multimeric mode. This will include self-assembling polyphosphazene macromolecules that have been tested clinically. 4. Immunological characterization of candidate E2 and E1E2 vaccines from Aims 1-3. Immunogenicity testing in mice will identify optimal combinations of HCV envelope immunogens, protein expression system, and immunoadjuvant formulations. Leading candidates will be tested for immunogenicity in macaques. These studies will lead to a rationally designed vaccine candidate to induce broadly Vn antibodies at sufficient titers to prevent HCV infection of different genotypes and subtypes.